Currency is the exposure most often hedged by accident and least often decided on purpose. An investor who buys foreign equities has taken two positions, one in the asset and one in the currency, and the second was not chosen. The professional question is not "should I hedge this trade" but "what is this mandate's hedge ratio, who set it, and when is it reviewed".
The spectrum has four practical points and each is a defensible policy. Unhedged keeps the currency exposure and is defensible where the currency is negatively correlated with the asset, which is the historical pattern for a dollar-based investor holding certain developed markets. Fully hedged removes it and is the default for fixed income, where currency volatility can exceed the entire return of the asset. Half-hedged is the honest position of an investor who does not have a currency view and does not want to be maximally wrong in either direction, and it minimises regret rather than variance. Over-hedged, a ratio above one, is a currency view expressed through the hedge, and it should be recognised and reported as an active position rather than as risk management.
Whoever sets the ratio owns the decision. In an institutional fund it belongs in the investment policy statement, set by the board or the investment committee, reviewed annually, and delegated to the manager only within a stated band. A manager who moves from 50% to 100% hedged because the currency looks expensive has taken an active currency position, and the only question is whether the mandate permitted it.
Advertisement
How does a rolling currency hedge work?
The mechanics are a rolling short forward. A US fund holding a €40,000,000 European sleeve sells euros forward three months at a time, and at each roll the outgoing contract settles against spot while a new one is struck at spot plus the forward points. The points are the interest differential, not a forecast, which is the single most misunderstood number in corporate treasury. With dollar rates above euro rates, the euro sells at a forward premium against the dollar, so the hedger earns the points.
Worked Example 4 rolls four quarters and produces the decomposition that matters. Total hedge P&L, −$304,000. Carry component, +$816,000, being the sum of the forward points, known at each roll. Spot component, −$1,120,000, being the currency move the hedge gave up, unknowable at each roll. The two sum exactly to the total. The hedge did not "cost" the carry and it did not "cost" the spot move: it earned one and surrendered the other, and only one of them could be predicted.
What refinements does a hedging policy need?
Two refinements complete the picture. A proxy hedge uses a liquid currency correlated with an illiquid one, and it substitutes correlation risk for currency risk in exactly the way the minimum-variance hedge ratio describes, which means the effectiveness is ρ² and the residual volatility is σ√(1 − ρ²), both of which should be reported. A macro hedge goes further and hedges a currency not because it is in the portfolio but because it proxies a risk that is, which is a view and should be labelled one. And on rebalancing: a static hedge fixes the notional at inception, so an equity sleeve that rises 15% ends the quarter 13% under-hedged, while a dynamic hedge resets the notional at each roll or at a stated drift threshold, paying more in transaction costs to keep the ratio near target. The choice between them is the same trade-off as everywhere else here, tracking error against cost, and it belongs in the policy document rather than in a monthly decision.
Set the ratio while no position is open, or it will be set for you by whichever way the currency last moved.